On 06.09.2017 15:17, Borislav Petkov wrote: > On Wed, Sep 06, 2017 at 02:54:05PM +0200, David Hildenbrand wrote: >> Maybe go one step further and incorporate everything (+vls) into a >> single if statement? > > Or maybe simplify it even more by not even looking at vls. If the user > disables it, fine, if she enables it but the hw doesn't support it, it > will be set to false automatically. > > Or am I missing a case? > > --- > From: Borislav Petkov <[email protected]> > Date: Tue, 5 Sep 2017 18:59:55 +0200 > Subject: [PATCH] KVM: SVM: Do not issue virtual VMLOAD/VMSAVE > supported-message > MIME-Version: 1.0 > Content-Type: text/plain; charset=UTF-8 > Content-Transfer-Encoding: 8bit > > There's no need to issue that everytime during boot - we have the > /proc/cpuinfo flag for people and software to query. > > Also, simplify logic which verifies the vls chicken bit setting. > > Signed-off-by: Borislav Petkov <[email protected]> > Cc: David Hildenbrand <[email protected]> > Cc: Janakarajan Natarajan <[email protected]> > Cc: Paolo Bonzini <[email protected]> > Cc: Radim Krčmář <[email protected]> > --- > arch/x86/kvm/svm.c | 13 ++++--------- > 1 file changed, 4 insertions(+), 9 deletions(-) > > diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c > index 8dbd8dbc83eb..d3c481778d9c 100644 > --- a/arch/x86/kvm/svm.c > +++ b/arch/x86/kvm/svm.c > @@ -1098,15 +1098,10 @@ static __init int svm_hardware_setup(void) > } > } > > - if (vls) { > - if (!npt_enabled || > - !boot_cpu_has(X86_FEATURE_V_VMSAVE_VMLOAD) || > - !IS_ENABLED(CONFIG_X86_64)) { > - vls = false; > - } else { > - pr_info("Virtual VMLOAD VMSAVE supported\n"); > - } > - } > + if (!npt_enabled || > + !boot_cpu_has(X86_FEATURE_V_VMSAVE_VMLOAD) || > + !IS_ENABLED(CONFIG_X86_64)) > + vls = false; > > return 0; > >
had the same idea but was worried about runtime. but this is really only executed once, so Reviewed-by: David Hildenbrand <[email protected]> -- Thanks, David
